[megerian]

Richard - It seems like it may indeed by the cpu after all, but just for clarity, can you actually post the exact behavior? Since you said that when in the completely closed position, the top doesnt not respond at all to the rocker switch pushed. You also said that once open, the switch moves the top forward and back?

So - some questions:

When the top is completely stuck how are you opening it, are you using the screwdriver in the manual hole technique?
Once its open and you said it correctly starts moving, does the closing sequence complete normally? By closing sequence, I mean that the front edge of the top drops onto the windshield, which in turns activates the closing motors.
If thats all true, then once the top drops onto the windshield, you should still be able to use the rocker switch forward or backward, and it should still work, to either activate the closing motors, or start the top moving backward again.

I am just trying to determine where in the sequence everything stops working.